In this paper we study, from both a theoretical and an experimental perspective, algorithms and data structures to process queries that help in the detection of rare variations over time intervals that occur in time series. Our research is strongly motivated by applications in financial domain.

The traditional base state with amendments model usually use the time sequence when choosing the base state to manipulate and it is usually use the extension of time point to represent the time-slot retrieval, which ignores the adjacent feature of the time point in a certain time-slot. The concept of operated base state and operation times is introduce in this paper, in which a base state with ...

A classical test for the detection of level shifts in such weakly dependent data is the CUSUM test, which compares the partial sum of the first m observations to the sum of all observations for each candidate change-point m, and maximizes this statistic with respect to m after some appropriate scaling. Asymptotical critical values for the CUSUM test can be calculated from tables of the Kolmogor...

We propose efficient (“fast” and low memory consuming) algorithms for universalcoding-based prediction methods for real-valued time series. Previously, for such methods it was only proved that the prediction error is asymptotically minimal, and implementation complexity issues have not been considered at all. The provided experimental results demonstrate high precision of the proposed methods. ...

This study examines the relation between the bid-ask spread from the daily CRSP data and the bid-ask spread from the intraday TAQ data. We show that the CRSP-based spread is highly correlated with the TAQ-based spread across stocks using data from 1993 through 2009. The simple CRSP-based spread provides a better approximation of the TAQ-based spread than all other low-frequency liquidity measur...

Is public R&D spending complementary and thus “additional” to private R&D spending, or does it substitute for and tend to “crowd out” private R&D? Conflicting answers are given to this question. We survey the body of available econometric evidence accumulated over the past 35 years. A framework for analysis of the problem is developed to help organize and summarize the findings of econometric s...

We present a simple and general result that the sign of the variations or increments of uncorrelated times series are predictable with a remarkably high success probability of 75% for symmetric sign distributions. The origin of this paradoxical result is explained in details. We also present some tests on synthetic, financial and global temperature time series.

This paper presents a discriminating approach to detecting the existence of underlying determinism in speech phonemes using the surrogate data method. The discrimination is made using a statistical measurement of neighboring trajectory directions. This approach is experimentally verified with both deterministic and stochastic time series and then applied to speech phonemes from the TIMIT databa...

This article proposes knowledge-based short-time prediction methods for multivariate streaming time series, relying on the early recognition of local patterns. A parametric, well-interpretable model for such patterns is presented, along with an online, classification-based recognition procedure. Subsequently, two options are discussed to predict time series employing the fuzzified pattern knowl...

In this paper we introduce a simple bivariate integer-valued time series model with positively correlated geometric marginals based on the negative binomial thinning mechanism. Some properties of the model are considered. The unknown parameters of the model are estimated using the modified conditional least squares method. © 2011 Elsevier Ltd. All rights reserved.